Points to Consider for Replacement of Conventional Tests for Viruses with Next Generation Sequencing (NGS) Assays — PMDA Early Consideration Summary

In Plain Language

The document addresses whether next generation sequencing (NGS)—a set of technologies that reads genetic material—can take the place of established tests used to detect viruses in certain biological products. It asks developers to validate an NGS assay in relation to the particular test being replaced, showing that the assay can detect the viruses of concern and has suitable performance for its intended purpose. The choices involved in that work include which reference viruses and specimens to use, how to extract nucleic acids, which sequencing platform to use, and how to analyse the resulting data. The document also distinguishes between replacing tests for known, specific viruses and tests intended to detect unspecified viruses. Its purpose is to set out PMDA’s current regulatory considerations while international work on suitable procedures and validation approaches continues.

What This Guidance Covers

The document applies when an NGS assay is used instead of one of four types of virus test for biopharmaceuticals or regenerative medical products within the scope of ICH Q5A(R2), as named in the document: a nucleic acid amplification test (NAT assay), an antibody production test, an in vivo test for viruses, or an in vitro test for viruses. The antibody production tests discussed include hamster antibody production (HAP), mouse antibody production (MAP), and rat antibody production (RAP) tests.

The document does not apply to biopharmaceuticals or regenerative medical products outside the scope of ICH Q5A(R2), although it notes that its essential concept may also be applicable to them. It also discusses a different situation: an NGS assay might be part of a comprehensive virus-control strategy because of a product-specific issue, rather than replacing one particular test. In that situation, PMDA consultation may be available to discuss the justification on a case-by-case basis.

The document is not a description of one standard NGS method. It explains that NGS technologies use different platforms and that test methods are not standardized. The document presents points to consider for validation and for replacement decisions, rather than a single globally agreed procedure.

Why It Matters

NGS has become more practical for virus detection as the technology has advanced and costs have fallen. Its high processing capacity offers a way to examine genetic material for viral sequences, including in settings where developers are considering replacing conventional virus tests. However, the document stresses that NGS is a technology, not a single uniform assay: platforms, sample preparation, databases, and analysis software can differ. Those choices may affect what the assay can detect and how results should be interpreted.

There is also an important distinction between detecting a virus’s genetic material and detecting infectious virus particles. A finding of virus-derived nucleic acids does not necessarily mean that infectious virus is present. In addition, the document notes that globally agreed NGS assay procedures and validation studies for replacing conventional virus tests are not yet available. International groups and research teams are working on appropriate methods and evaluation. Against that background, PMDA describes its current considerations and notes that they may change as research and knowledge develop.

Key Recommendations

1. Design validation around the test being replaced. The validation should address whether the NGS assay can meet the expected performance standard for that particular test. The document points to the analytical target profile (ATP) described in ICH Q14 as a basis for setting the validation strategy in light of the study objective. It also cautions developers to consider whether additional validation factors are needed as technology and understanding advance.

2. Choose reference viruses and specimens to fit the assay’s purpose. Validation should establish the detection limit, specificity, and range (including the type of virus) of the target virus or viruses. Reference viruses should be selected according to whether the assay targets specific viral groups or aims to detect unspecified viruses. Specimen choice should likewise take account of the strengths and limitations of genomic, transcriptomic, and viromic approaches.

3. Evaluate sample preparation and sequencing as an integrated assay. Extraction efficiency can depend on the specimen and virus characteristics. Developers should establish an extraction system suitable for the target viruses and carefully optimize the kit protocol. Library preparation should be assessed for the NGS platform used, and library quality should be evaluated comprehensively by considering its ability to detect reference viruses.

4. Select the sequencing platform, database, and software with the intended detection task in mind. Short-read and long-read platforms have different characteristics that may affect detection. Database coverage and bioinformatics software performance are important, as are analysis parameters such as the number of reads. Developers should choose a database shown to perform appropriately for the test objective and manage database and software versions. When a version changes, its impact should be evaluated and the analytical procedure revalidated as necessary.

5. Set in advance how detected viral nucleic acids will be handled. Because detection of viral genetic material does not necessarily show that infectious particles are present, protocols for dealing with a finding should be predetermined in view of the test’s position in the overall process. Further investigation may consider infectivity, the length and coverage of detected sequences, whether the nucleic acid is DNA or RNA, the NGS platform, and a possible contamination route. If a conclusion is based on additional investigation, the presence or absence of infectious viruses should be evaluated by integrating relevant factors.

6. For replacement of NAT assays, demonstrate detection of the specific viruses of concern. The NGS assay must be shown to detect the target specific viruses. The document says a head-to-head comparison of detection sensitivity may have limited practical significance. Instead, developers should collect information to explain that the NGS assay is expected to have virus-detection ability comparable to the NAT assay and can detect the specific viruses with sufficient sensitivity to meet expected performance criteria.

7. For replacement of antibody production tests, demonstrate detection of the specified viruses. These tests use an animal’s antibody response as an indicator of particular viruses. If an NGS assay or NAT assay is used instead, it must be demonstrated to detect the viruses listed in Table 3 of ICH Q5A(R2). A head-to-head comparison is not required because the approaches use fundamentally different platforms. Developers should collect information showing sufficient sensitivity for the viruses of concern and that expected performance criteria are met.

8. For replacement of in vivo tests, show adequate coverage alongside an in vitro virus test. In vivo tests are intended to detect unspecified adventitious viruses and use different principles from NGS, so a head-to-head comparison is impractical. The document says an NGS assay may be accepted as an alternative on the assumption that an in vitro test for viruses is performed, if the NGS assay is shown to have detection coverage wide enough to complement that in vitro test. It also notes that replacement may have limited practical significance for well-characterized rodent cells, such as CHO cells, which are not required to undergo in vivo tests for viruses.

9. Do not treat full replacement of in vitro tests as ready for routine use. The document notes that in vitro tests have historically played a central role in detecting adventitious viruses in biotechnology products and that international agreement on strategies for validating complete replacement with NGS has not been reached. It considers it premature, at present, to use NGS assays in place of in vitro tests for viruses. For now, developers should pay close attention to international activities and make efforts to collect information. If NGS is used as part of an integrated virus-safety strategy throughout manufacturing, but not in place of an in vitro test, its appropriateness would be judged for the specific product and consultation with the applicable reviewing office should be considered.

Practical Implications

Interpretation: The document’s approach makes the intended use of the assay the starting point for development. A method intended to detect a defined group of viruses raises different validation questions from a method intended to find a broad range of unspecified viruses. Sponsors will need to connect the assay’s purpose to decisions about reference viruses, specimen type, sample preparation, sequencing technology, data resources, and result handling. The document does not supply a universal validation recipe; the case for performance depends on the test being replaced and the assay’s objective.

Interpretation: Planning should account for the full chain from specimen to conclusion. For example, a platform may produce sequence data, but detection depends on the material that enters the assay, how it is prepared, and how software and databases identify sequences. Version management is therefore part of maintaining the validated procedure, not merely an information-technology detail. If a database or software version changes, the document calls for assessing the impact and repeating validation as necessary.

Interpretation: Teams should distinguish an analytical finding from a decision about infectious risk. A positive sequence finding may need further investigation under a protocol established in advance. The considerations listed by PMDA—such as infectivity testing, sequence length and coverage, nucleic-acid type, platform, and contamination route—can inform that investigation, but the document does not present any one factor as a standalone decision rule.

For regulatory and study planning teams, replacement of each conventional test warrants a separate rationale. The document describes different expectations for NAT assays, antibody production tests, in vivo tests, and in vitro tests. In particular, the current position on replacing in vitro testing is more cautious than the discussion of the other test types. A product-specific integrated virus-safety approach that supplements rather than replaces an in vitro test is also treated separately, with appropriateness assessed for the product.

Technical Details

The document describes NGS as sequencing genomic nucleic acids at high throughput. In common short-read sequencing, many short DNA fragments are read in parallel and computer software combines the resulting reads to reconstruct the sequence. Long-read sequencing can read DNA fragments longer than 1,000 bases in a single run. The document contrasts these platform groups: short-read platforms are expected to provide highly accurate base sequences but may have difficulty with some regions, including repetitive sequences, while long-read platforms can handle such regions but may have higher error rates when reading many bases in one pass.

For validation, the document identifies several linked choices:

• Reference viruses: For detection of specific viral groups, choose viruses with physical, chemical, and genomic properties similar to the target analytes. For comprehensive detection of unspecified viruses, select multiple viruses with different properties to cover all virus types. The document mentions a seven-virus candidate panel proposed in the cited work, but says the panel may change and developers should explain why the selected reference viruses suit the test objective.
• Specimen type: Genomic analysis uses genomic nucleic acids extracted from cells; transcriptomic analysis uses messenger RNA (mRNA) extracted from cells; and viromic analysis uses nucleic acids prepared from cell-culture supernatants. Transcriptomic analysis may suit viruses with ongoing mRNA replication but may be less sensitive to latent viruses. Viromic analysis may suit viruses shed into culture medium but may be less sensitive to latent viruses or viruses that spread cell-to-cell without budding particles. Genomic analysis is less likely to have those biases, but may misidentify virus-like sequences in chromosomal DNA. The selected specimen should be justified in view of these advantages and limitations.
• Manufacturing-related material: Specimens may contain substantial cell-derived nucleic acids, including endogenous virus-like sequences. Material from processes involving viral vectors, or products containing viruses or vectors as active ingredients, may also contain high concentrations of their nucleic acids. The document says validation may use specimens collected from the actual manufacturing process and spiked with reference viruses, so the assay is assessed in the presence of relevant background material.
• Extraction and library preparation: Extraction efficiency depends on specimen type and virus characteristics. Column- or magnetic-bead-based kits are commonly used, but details such as specimen and nucleic-acid amounts, whether carrier DNA or RNA is added, and the final elution volume can affect extraction. Library preparation depends on the selected NGS platform and should be evaluated in that platform’s context. The ability to detect reference viruses is part of the comprehensive assessment of library quality.
• Database, software, and parameters: The document identifies the Reference Viral Database (RVDB) as a commonly used resource with extensive viral coverage. Broad coverage can also include viruses without documented human pathogenicity and endogenous virus-like sequences, creating “noise” that may compromise detection of viruses that should be detected. A database limited to viruses infectious to humans, such as “Human viruses” of Virus-Host DB, may be used to address that concern. The chosen database should be shown to perform in line with the test objective. Software parameters, including the number of reads, should be determined carefully.

The document classifies an NGS assay as a limit test under ICH Q2(R2), and says that validation characteristics should be specified in view of that document. It does not give a single set of performance thresholds for all assays. Instead, it emphasizes defining the validation strategy in relation to the intended purpose and expected performance of the conventional test being replaced.

Who Should Read It

This document is relevant to developers and marketing authorization holders considering NGS for virus testing of biopharmaceuticals or regenerative medical products within the stated scope. It is also useful to analytical and virology scientists planning assay development and validation; bioinformatics teams responsible for databases, software, and analysis parameters; quality and manufacturing staff selecting representative specimens; and regulatory teams preparing a justification for replacement or an integrated virus-safety strategy. Investigators and reviewers involved in evaluating assay findings may also benefit from its distinction between viral nucleic-acid detection and evidence of infectious virus.

Official PMDA Source

• Official title: Points to Consider for Replacement of Conventional Tests for Viruses with Next Generation Sequencing (NGS) Assays (Early Consideration)
• Issuing body: Pharmaceuticals and Medical Devices Agency (PMDA), Center for Product Evaluation
• Date printed on the document: January 22, 2026
• Official PMDA PDF: View the PMDA document
• PMDA list page: View PMDA Early Consideration documents

Source: Pharmaceuticals and Medical Devices Agency (PMDA) website, PMDA Early Consideration documents. This page was created by editing and summarising the PMDA document.

Consult the official document for the full text; this summary does not replace it. PMDA provides the English version for reference, and the Japanese original prevails if the two versions differ. PMDA describes Early Consideration documents as reference information reflecting its views at the time of publication; those views may change as scientific knowledge develops.

Related Clinical Biostats Resources

• All Clinical Biostats tutorials — Browse other educational tutorials, including summaries of PMDA Early Consideration documents.
• Statistical Calculators — Access calculators for statistical tasks that may support analysis planning and interpretation.